Nelder-Mead Simplex Channel Estimation for the RF-DNA Fingerprinting of OFDM Transmitters Under Rayleigh Fading Conditions

The Internet of Things (IoT) is a collection connected devices capable interacting with the physical world and computer systems. It estimated that IoT will consist more than seventy five billion by year 2025. In addition to sheer numbers, need for security exacerbated fact many edge employ weak no encryption communication link. has been almost 70% use form encryption. Previous research suggested Specific Emitter Identification (SEI), layer technique, as means augmenting bit-level mechanisms such Radio Frequency-Distinct Native Attributes (RF-DNA) fingerprinting an SEI technique demonstrated success in discriminating radios operating within noise only channel. This work extends RF-DNA discrimination under Rayleigh fading conditions through Nelder-Mead (N-M) simplex-based channel estimator. N-M estimator estimates multipath directly from received waveform; thus, eliminating demodulation required when using constellation-based estimators. proves superior three alternative waveform-based estimation approaches increasing paths/reflections decreasing Signal-to-Noise Ratio (SNR). performance maximized assessment of: (i) fingerprints generated magnitude versus phase representation Gabor transform's coefficients, (ii) statistic-based classifier neural network-based classifier, (iii) size patch used subdivide Gabor-based time-frequency response prior calculation fingerprint features. resulting process achieves average percent correct classification 92.3% or greater channels consisting two, three, reflections/paths at SNR≥15 dB.